Effects of warm-up before eccentric exercise
on indirect markers of muscle damage
RACHEL K. EVANS, KENNETH L. KNIGHT, DAVID O. DRAPER, and ALLEN C. PARCELL
College of Health and Human Performance, Brigham Young University, Provo, UT
ABSTRACT
EVANS, R. K., K. L. KNIGHT, D. O. DRAPER, and A. C. PARCELL. Effects of warm-up before eccentric exercise on indirect
markers of muscle damage. Med. Sci. Sports Exerc., Vol. 34, No. 12, pp. 1892–1899, 2002. Purpose: To test whether active and passive
warm-up conducted before eccentric exercise attenuates clinical markers of muscle damage. Methods: Untrained subjects were exposed
to one of five conditions: low-heat passive warm-up (N ⫽ 10), high-heat passive warm-up (N ⫽ 4), or active warm-up (N ⫽ 9),
preceding eccentric exercise; eccentric exercise without warm-up (N ⫽ 10); or high-heat passive warm-up without eccentric exercise
(N ⫽ 10). Passive warm-up of the elbow flexors was achieved using pulsed short-wave diathermy, and active warm-up was achieved
by concentric contraction. Creatine kinase (CK) activity, strength, range of motion, swelling, and muscle soreness were observed before
treatment (baseline) and 24, 48, 72, and 168 h after treatment. Results: High-heat passive warm-up without eccentric exercise did not
affect any marker of muscle damage and was used as our control group. Markers of muscle damage were not different between groups
that did or did not conduct warm-up before eccentric exercise. The active warm-up and eccentric groups exhibited a greater
circumferential increase than controls (P ⬍ 0.0002), however, that was not observed after passive warm-up. Additionally, the active
warm-up group exhibited a greater CK response than controls at 72 h (P ⬍ 0.05). The high-heat passive warm-up before eccentric
exercise group exhibited significant change from controls at the least number of time points, but due to a small sample size (N ⫽ 4),
these data should be viewed as preliminary. Conclusion: Our observations suggest that passive warm-up performed before eccentric
exercise may be more beneficial than active warm-up or no warm-up in attenuating swelling but does not prevent, attenuate, or resolve
more quickly the other clinical symptoms of eccentric muscle damage as produced in this study. Key Words: ACTIVE WARM-UP,
PASSIVE WARM-UP, DIATHERMY, MUSCLE TEMPERATURE, MUSCLE SORENESS
S
manifestations of damage that include high plasma creatine
kinase (CK) activity, strength loss, range of motion loss,
swelling, and delayed onset muscle soreness (27). Although
most individuals have experienced this type of injury, there
is no evidence that any specific intervention consistently
prevents, attenuates, or treats the clinical symptoms of eccentric strain injury.
Warm-up of muscle is routinely recommended as a way
to prevent strain injury (26,28), despite a lack of scientific
evidence to support the clinical efficacy of warm-up in
injury prevention. There is, however, evidence to suggest
that elevating muscle temperature might decrease strain
injury through changes in muscle tissue extensibility, by
allowing increased stretch to occur before the onset of
damage in both tendon (13,31) and muscle tissue (21,26,30).
Increased muscle temperature can be achieved actively,
through exercise, or passively, through the use of heat
modalities. Temperature elevations observed after 15 s to 15
min of active muscle contraction range from 1.0 to 3.3°C
(24,26,32). Modalities used in sports medicine clinics to
passively raise tissue temperature include ultrasound, shortwave diathermy, and warm-water immersion, all of which
have shown to raise deep muscle temperature between 3.4
and 3.8°C after 10- to 30-min treatments (4 – 6). It is unclear,
however, what degree of temperature elevation is necessary
to increase tissue extensibility to levels that may prevent
injury. An increase in muscle extensibility has been observed after a 1°C elevation (26) and in tendon after a 3– 4°C
elevation (12). Other researchers have recommended that
tretch-induced injury to muscle, or muscle strain, is
one of the most common injuries experienced by a
recreationally active population (19). Muscle strain
can occur when muscle is stretched while in a passive state,
or when the muscle is actively contracting (7). The latter is
a lengthening, or eccentric contraction, and has been well
established as the most damaging type of muscle action
(2,20).
During unaccustomed eccentric exercise, damage results
when strain exceeds the limits of the connections between
noncontractile elements in the cytoskeletal and extracellular
matrices, such as collagen and other structural proteins
(18,29). This structural damage results in microscopic disruption of muscle fiber elements (3,14,17,20), the extracellular connective tissue matrix (2,9,29), or a combination of
these structures and initiates a series of reactions that lead to
loss of cellular integrity, disrupted calcium homeostasis, and
cell necrosis (1). Ultimately, this process results in clinical
Address for correspondence: Rachel K. Evans, Military Performance Division,
United States Army Research Institute of Environmental Medicine, 42 Kansas
Street, Natick, MA 01760; E-mail: rachel.evans@na.amedd.army.mil.
Submitted for publication February 2002.
Accepted for publication August 2002.
0195-9131/02/3412-1892/$3.00/0
MEDICINE & SCIENCE IN SPORTS & EXERCISE®
Copyright © 2002 by the American College of Sports Medicine
DOI: 10.1249/01.MSS.0000038895.14935.C8
1892
tissue reach a specific temperature for extensibility to occur
(13,32). Most studies have looked at the degree of extensibility that occurs prior to tissue rupture, however, rather than
onset of strain injury.
Recently, Nosaka and Clarkson (22) reported that an
active warm-up of 100 concentric contractions performed
just before an eccentric exercise bout decreased indicators of
muscle damage. They concluded that exercise before the
eccentric bout may have warmed up the muscle, minimizing
damage. We are not aware of any studies that have assessed
the effects of passive warm-up before eccentric exercise. In
addition, studies are lacking that have compared the effects
of passive and active warm-up and their role in preventing
strain injury to muscle.
This study was designed, therefore, to test the effects of
warm-up on muscle damage resulting from eccentric exercise and to compare the effects of passive versus active
warm-up. Additionally, we explored the effects of a highheat passive warm-up before eccentric exercise. We hypothesized that warm-up before eccentric exercise would attenuate indirect markers of eccentric muscle damage and that
active and passive warm-up methods designed to elevate
muscle temperature by similar levels would have similar
effects. We also hypothesized that passive warm-up of
3– 4°C would be more effective than passive warm-up of
1°C in attenuating damage.
22.4 ⫾ 5.9 yr). Subjects were excluded if they had experienced recent illness, were on medications that might interfere with study results, or had participated in any resistancetraining program during the 3-month period before the
study. Written informed consent was obtained from all
subjects in accordance with the Institutional Review Board
for Research with Human Participants. Volunteers agreed to
refrain from any new or strenuous physical activity while
participating in the study.
Thirty-nine subjects were randomly assigned to one of
four treatment groups (CONTROL (N ⫽ 10; 3 male, 7
female), ECC (N ⫽ 10; 2 male, 8 female), PW⫺LO⫹ECC
(N ⫽ 10; 4 male, 6 female)), or AW⫺LO⫹ECC (N ⫽ 9; 4
male, 5 female). Due to the remote possibility that elevated
muscle temperature would significantly increase eccentric
muscle damage (33), we limited our PW⫺HI⫹ECC treatment to a small group of four volunteers (3 male, 1 female).
Each subject reported to the research lab on six separate
days. On day 1, subjects were familiarized with the study
methods, and baseline measurements of the dependent variables were obtained. On day 2, subjects were treated with
their assigned experimental condition. Subjects then reported to the lab at 24, 48, 72, and 168 h (days 3– 6) after
treatment, where postexercise measurements were taken,
identical to those taken at baseline. The dominant arm was
tested in all cases.
METHODS
Experimental Procedure
Experimental Design
Passive warm-up. Passive warm-up was achieved utilizing a Magnatherm SSP (1000 SS) pulsed short-wave
diathermy unit (International Medical Electronics, Ltd.,
Kansas City, MO) with a 24-cm diameter induction applicator operating at 27.12 MHz. Heat delivery is controlled
using 12 independently adjustable power (pulse train width)
and rate (pulse·s⫺1) settings, where 12 represents 100% of
maximal capability, and 6 represents 50% of maximal capability. To passively heat the elbow flexors, the subject was
placed supine, the shoulder abducted approximately 40°,
and the forearm supported at the wrist to place the elbow in
slight flexion. The applicator head was then positioned
directly over the muscle belly of the biceps brachii. For the
PW⫺LO⫹ECC condition, heat was applied to the muscle
for 10 min at a power and rate setting of 5 and 5 (5/5),
respectively. For the PW⫺HI⫹ECC group, heat was applied for a period of 10 min at a 12/12 setting. The high-heat
passive warm-up alone group (CONTROL) was also treated
for 10 min at a 12/12 setting. Pilot work conducted in our
laboratory estimated that the temperature rise of the elbow
flexor muscles after 10-min treatments at settings of 5/5 and
12/12, as determined by thermocouple needle microprobes
placed in the biceps brachii to a depth of 2 cm, was approximately 1° and 3.5°C, respectively. These muscle temperature increases are similar to those found in a previous
investigation assessing change in deep muscle temperature
at a depth of 3 cm (5).
Active warm-up. The BIODEX dynamometer (Biodex
Medical Systems, Shirley, NY) was used to actively
A 5 ⫻ 5 factorial with repeated measures on one factor
guided this study. The two independent variables were time
(repeated measures at baseline, 24, 48, 72, and 168 h) and
treatment group, the latter of which included high-heat
passive warm-up without eccentric exercise (CONTROL),
eccentric exercise without warm-up (ECC), low-heat
passive warm-up preceding eccentric exercise (PW⫺
LO⫹ECC), high-heat passive warm-up preceding eccentric
exercise (PW⫺HI⫹ECC), or active warm-up preceding eccentric exercise (AW⫺LO⫹ECC). Pilot work determined
that passive warm-up, achieved by applying pulsed shortwave diathermy to the elbow flexors, resulted in temperature elevation of the biceps brachii by approximately 1°C
(PW⫺LO) or 3.5°C (PW⫺HI), depending on the setting
used. Active warm-up was achieved by performing 100
concentric contractions of the elbow flexors and was found
to elevate temperature of the biceps brachii by approximately 1°C (AW⫺LO). Treatment effects were assessed by
observing changes in CK activity, isometric strength
(MVC), resting elbow extension, active elbow flexion, circumference at the biceps muscle belly and distal biceps
tendon region, and muscle soreness with extension and
flexion.
Subjects
Forty-three healthy male (N ⫽ 16) and female (N ⫽ 27)
college students volunteered to participate in this study (age
WARM-UP BEFORE ECCENTRIC EXERCISE
Medicine & Science in Sports & Exercise姞
1893
warm-up the elbow flexors, without inducing fatigue, using
a previously established method (22). After being secured in
the test position, the subject moved the elbow joint smoothly
from an extended (170°) to a flexed (50°) position and back
100 times, completing each repetition in 2 s. The BIODEX
was set at 120°·s⫺1 to minimize the load and to ensure that
fatigue-provoking forces were not generated. We determined that that this active warm-up protocol elevates muscle temperature of the elbow flexor muscle group, as determined by pilot work identical to that used to assess changes
in deep muscle temperature of the biceps brachii after passive warm-up, by approximately 1°C.
Eccentric exercise. The passive mode of a computerinterfaced dynamometer (BIODEX System 3) was used to
induce muscle damage (29), and was conducted immediately (within 10 –15 s) after any warm-up procedure. Subjects were seated with the arm supported and were stabilized
at the waist and chest. Starting with the elbow flexed to 50°
and ending at an angle of 170°, each subject performed 50
maximal eccentric movements of the elbow flexors at
120°·s⫺1. During each movement, subjects were verbally
encouraged to produce a maximal effort to resist the ability
of the dynamometer to extend the elbow. Subjects were
given a 10-s rest between each contraction, during which
time the dynamometer arm returned passively to the starting
position.
Blood sampling and analysis. Approximately 5 mL
of blood were taken from an antecubital vein using sterile
venipuncture techniques. Blood was collected in EDTAcoated tubes (Becton Dickinson, Franklin Lakes, NJ) and
immediately centrifuged for 10 min. Plasma samples were
separated and stored at ⫺80°C until analysis. Plasma CK
activity was measured in duplicate using an enzymatic assay
kit (Sigma 47UV, St. Louis, MO).
Strength. Maximal voluntary isometric contraction of
the elbow flexors was measured using the BIODEX dynamometer. After seating the subject in the test position with
the elbow flexed to an angle of 90°, two 3-s maximal
isometric contractions of the elbow flexor muscle group
were performed, with a 1-min rest between trials. The highest value of the two trials (in N·m) was used to represent
MVC.
Range of motion. Measurements of elbow range of
motion were taken using a 30-cm clear plastic goniometer
(Bissell Healthcare Corp., Brookfield, IL). The anatomic
reference points for goniometer placement were the deltoid
insertion, the lateral humeral epicondyle, and the ulnar
styloid process. These points were marked with semipermanent ink when baseline measurements were taken and were
maintained during the study. The relaxed angle of the elbow
joint reflects changes in muscle stiffness of the elbow flexors and was measured with the subject standing with the arm
relaxed. Flexion was measured after asking the subject to
bend the elbow as much as possible while reaching for the
ipsilateral shoulder.
Circumference. Swelling was assessed by circumference measurements taken at the biceps tendon region and
the biceps muscle belly (1 cm and 6 cm proximal to the
1894
Official Journal of the American College of Sports Medicine
humeral epicondyles, respectively) by using a standard tape
measure. Measurement sites were kept consistent by repeatedly marking the arm with semipermanent ink. With the
subject standing and the upper extremity in a neutral, relaxed position, three measurements were taken at each point
and the average recorded as the score (in cm).
Muscle soreness. Subjects rated muscle soreness on a
100-mm visual analog scale (VAS), with the far-left end
point representing no pain and the far-right end point representing very sore muscles. Standing with the upper limb
relaxed and the forearm fully supinated, subjects actively
extended the elbow. They then placed a mark on the VAS
that represented the soreness experienced in the elbow
flexor region of the arm during the motion. In the same
manner, they rated the soreness experienced during full
flexion of the elbow.
Statistical Analyses
The data were analyzed using SAS statistical software
(Statistical Analysis System Institute, Inc., Cary, NC) to
conduct a repeated measures analysis of variance (ANOVA)
between the five treatment groups for each dependent variable at five time points. If significant main effects or interactions were found, Tukey post hoc tests were used to
determine where differences existed. Statistical significance
was set at P ⬍ 0.05.
RESULTS
There were no differences in baseline measurements between groups for any dependent variable (Table 1). Highheat passive warm-up without eccentric exercise (CONTROL) had no significant effect on any dependent variable
at any time (Table 1) and was used as the control group.
Although significant differences were observed in how the
experimental groups differed from the control group, we
observed no significant difference in markers of muscle
damage between the groups that conducted warm-up before
eccentric exercise and the eccentric exercise without
warm-up group.
Plasma CK activity. Creatine kinase exhibited large
intersubject variability (Table 1). For all groups combined,
CK levels exceeded baseline values only at 72 h (F4,152 ⫽
7.86, P ⬍ 0.001, Table 1), at which point values were higher
than at 24, 48, and 168 h. There was no overall treatment
effect (F4,38 ⫽ 1.90, P ⫽ 0.13); however, post hoc analysis
of a significant time-by-treatment interaction (F16,152 ⫽
1.88, P ⫽ 0.03) revealed that CK activity was greater in the
AW⫺LO⫹ECC than in the CONTROL group at 72 h (Fig. 1).
Strength. Strength was analyzed as the percent change
from baseline MVC to account for gender differences observed at baseline. Strength changed significantly over time
(F4,152 ⫽ 44.79, P ⬍ 0.001) and was significantly less than
control values at 24, 48, 72, and 168 h for all experimental
groups (Table 1). Experimental groups were not different
from each other but were collectively different than the
control group (F4,38 ⫽ 10.82, P ⬍ 0.001) and exhibited a
http://www.acsm-msse.org
TABLE 1. Measurement of indirect markers of muscle damage (mean ⫾ SD).
Baseline
24 h
48 h
72 h
168 h
86.0 ⫾ 96.5
66.0 ⫾ 60.4
50.2 ⫾ 37.1
89.5 ⫾ 37.4
74.7 ⫾ 100.9
126.8 ⫾ 185.3
124.9 ⫾ 132.3
116.6 ⫾ 145.2
213.5 ⫾ 86.2
415.8 ⫾ 901.4
100.2 ⫾ 121.2
569.4 ⫾ 1250.7
179.0 ⫾ 215.6
449.7 ⫾ 757.5
1262.2 ⫾ 1822.1
102.7 ⫾ 108.9
2109.1 ⫾ 3331.3
970.9 ⫾ 503.7
2096.7 ⫾ 983.8
5199.4 ⫾ 8183.3
63.1 ⫾ 29.0
1245.3 ⫾ 1304.1
654.6 ⫾ 607.6
780.3 ⫾ 1032.2
1074.2 ⫾ 918.8
47.2 ⫾ 20.1
41.2 ⫾ 18.6
50.2 ⫾ 17.1
58.9 ⫾ 14.2
52.0 ⫾ 25.4
48.7 ⫾ 21.3
24.2 ⫾ 11.9
32.0 ⫾ 19.0
37.7 ⫾ 17.1
33.4 ⫾ 15.9
48.4 ⫾ 21.3
24.5 ⫾ 11.5
38.2 ⫾ 24.2
37.7 ⫾ 11.6
33.7 ⫾ 13.0
47.9 ⫾ 20.8
24.5 ⫾ 12.3
37.7 ⫾ 22.4
41.3 ⫾ 19.1
33.4 ⫾ 12.3
48.0 ⫾ 21.2
30.2 ⫾ 15.3
42.3 ⫾ 20.9
47.1 ⫾ 17.3
39.1 ⫾ 18.0
19.5 ⫾ 6.6
21.3 ⫾ 8.6
17.5 ⫾ 8.5
21.8 ⫾ 3.5
26.0 ⫾ 4.8
19.1 ⫾ 7.0
30.3 ⫾ 7.0
23.5 ⫾ 7.8
27.3 ⫾ 4.0
35.1 ⫾ 13.3
18.6 ⫾ 6.4
34.1 ⫾ 9.9
24.5 ⫾ 8.0
27.8 ⫾ 2.5
35.2 ⫾ 14.0
19.3 ⫾ 6.9
33.6 ⫾ 10.8
25.7 ⫾ 10.0
25.3 ⫾ 1.9
33.3 ⫾ 12.8
19.3 ⫾ 7.4
25.0 ⫾ 5.4
20.1 ⫾ 7.2
22.8 ⫾ 4.0
27.3 ⫾ 10.0
142.7 ⫾ 5.2
145.5 ⫾ 3.6
143.5 ⫾ 8.4
144.8 ⫾ 4.5
144.7 ⫾ 3.9
143.3 ⫾ 6.1
140.8 ⫾ 4.1
135.4 ⫾ 9.6
138.5 ⫾ 12.2
140.2 ⫾ 5.1
143.6 ⫾ 7.2
141.5 ⫾ 5.4
137.0 ⫾ 8.1
140.3 ⫾ 7.6
138.7 ⫾ 6.6
143.7 ⫾ 6.7
140.3 ⫾ 4.9
136.2 ⫾ 9.7
140.0 ⫾ 12.1
139.6 ⫾ 7.4
143.7 ⫾ 5.5
143.9 ⫾ 4.2
138.8 ⫾ 7.7
143.5 ⫾ 7.1
142.6 ⫾ 5.0
29.1 ⫾ 3.4
27.3 ⫾ 2.3
27.3 ⫾ 2.5
28.9 ⫾ 4.4
27.9 ⫾ 2.6
29.0 ⫾ 3.5
27.8 ⫾ 2.1
27.5 ⫾ 2.4
29.5 ⫾ 4.7
28.7 ⫾ 2.7
29.0 ⫾ 3.2
28.1 ⫾ 2.4
27.7 ⫾ 2.3
29.4 ⫾ 4.8
28.8 ⫾ 2.9
29.0 ⫾ 3.3
28.4 ⫾ 2.9
28.0 ⫾ 2.2
30.0 ⫾ 5.7
29.2 ⫾ 3.0
29.1 ⫾ 3.4
28.1 ⫾ 2.5
27.8 ⫾ 2.1
29.9 ⫾ 5.5
28.9 ⫾ 2.9
26.6 ⫾ 2.5
25.1 ⫾ 1.8
24.9 ⫾ 2.0
25.9 ⫾ 3.1
25.3 ⫾ 2.7
26.5 ⫾ 2.5
25.4 ⫾ 1.7
25.0 ⫾ 1.9
26.3 ⫾ 3.0
25.7 ⫾ 2.8
26.4 ⫾ 2.5
25.6 ⫾ 1.8
25.2 ⫾ 1.8
26.2 ⫾ 3.2
26.2 ⫾ 3.2
26.5 ⫾ 2.5
26.1 ⫾ 2.4
25.5 ⫾ 1.7
26.6 ⫾ 4.1
26.6 ⫾ 3.5
26.5 ⫾ 2.6
25.9 ⫾ 2.0
25.4 ⫾ 1.5
26.8 ⫾ 4.3
26.3 ⫾ 3.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
28.6 ⫾ 13.4
35.3 ⫾ 25.5
24.8 ⫾ 3.7
40.1 ⫾ 21.7
0.0 ⫾ 0.0
44.0 ⫾ 23.2
32.9 ⫾ 22.4
31.8 ⫾ 27.3
41.9 ⫾ 22.0
0.0 ⫾ 0.0
44.5 ⫾ 32.1
41.2 ⫾ 33.0
19.5 ⫾ 25.9
28.3 ⫾ 19.7
0.0 ⫾ 0.0
1.4 ⫾ 2.4
1.1 ⫾ 1.9
3.5 ⫾ 7.0
7.1 ⫾ 13.5
0.0 ⫾ 0.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
0.0 ⫾ 0.0
31.6 ⫾ 18.2
43.0 ⫾ 32.0
19.8 ⫾ 9.5
29.3 ⫾ 23.4
0.0 ⫾ 0.0
48.8 ⫾ 21.4
44.1 ⫾ 24.2
36.3 ⫾ 15.2
38.8 ⫾ 25.6
0.0 ⫾ 0.0
44.4 ⫾ 28.9
40.3 ⫾ 30.2
21.5 ⫾ 17.2
28.2 ⫾ 17.9
0.0 ⫾ 0.0
0.8 ⫾ 1.6
0.5 ⫾ 1.1
1.8 ⫾ 3.5
1.3 ⫾ 2.1
⫺1
Creatine kinase (U䡠L )
CONTROL
ECC
PW⫺LO⫹ECC
PW⫺HI⫹ECC
AW⫺LO⫹ECC
MVC (N䡠m)
CONTROL
ECC
PW⫺LO⫹ECC
PW⫺HI⫹ECC
AW⫺LO⫹ECC
Resting extension (deg)
CONTROL
ECC
PW⫺LO⫹ECC
PW⫺HI⫹ECC
AW⫺LO⫹ECC
Active flexion (deg)
CONTROL
ECC
PW⫺LO⫹ECC
PW⫺HI⫹ECC
AW⫺LO⫹ECC
Prox circumference (cm)
CONTROL
ECC
PW⫺LO⫹ECC
PW⫺HI⫹ECC
AW⫺LO⫹ECC
Distal circumference (cm)
CONTROL
ECC
PW⫺LO⫹ECC
PW⫺HI⫹ECC
AW⫺LO⫹ECC
Extension soreness (mm)
CONTROL
ECC
PW⫺LO⫹ECC
PW⫺HI⫹ECC
AW⫺LO⫹ECC
Flexion soreness (mm)
CONTROL
ECC
PW⫺LO⫹ECC
PW⫺HI⫹ECC
AW⫺LO⫹ECC
CONTROL, passive warm-up (high) without exercise; ECC, eccentric exercise without warm-up; PW⫺LO⫹ECC, passive warm-up (low) preceding eccentric exercise; PW⫺HI⫹ECC,
passive warm-up (high) preceding eccentric exercise; AW⫺LO⫹ECC, active warm-up preceding eccentric exercise.
38.2% loss of strength at 24 h compared with baseline
values (P ⬍ 0.05, Fig. 2). Significant strength loss was
evident throughout the course of the study and, although
significantly improved at 168 h when compared with the
24-, 48-, and 72-h measures (P ⬍ 0.05), was only 78 ⫾ 16%
of control values.
Range of motion. Resting extension and active flexion
measurements are presented in Table 1. Extension and flexion loss was significant at 24, 48, and 72 h after eccentric
exercise with or without warm-up (F4,152 ⫽ 20.57 and
19.33, P ⱖ 0.0001) and continued to be evident for flexion
at 168 h. All groups conducting eccentric exercise lost a
significant degree of extension when compared with controls (F4,38 ⫽ 2.97, P ⬍ 0.03), whereas only the
PW⫺LO⫹ECC and AW⫺LO⫹ECC groups lost more flexion than controls (F4,38 ⫽ 5.02, P ⫽ 0.002). Post hoc
analysis of significant time-by-treatment interactions
(F16,152 ⫽ 2.59 and 2.55, P ⫽ 0.001) for both extension and
WARM-UP BEFORE ECCENTRIC EXERCISE
flexion revealed significant differences from control values
for all experimental groups at one or more time points. The
ECC group exhibited significant extension loss at the greatest number of time points; the PW⫺LO⫹ECC exhibited
significant flexion loss at the greatest number of time points;
and the PW⫺HI⫹ECC group exhibited significant change
at the least number of time points for both extension and
flexion (Fig. 3).
Circumference. Circumferential measurements of both
the proximal and distal arm are presented in Table 1. Due to
significant differences in baseline girth measurements between male and female subjects, we analyzed the percent
change in circumference compared to baseline. A significant
increase over control values was observed beginning at 24 h
proximally (F4,152 ⫽ 27.79, P ⬍ 0.0001) and 48 h distally
(F4,152 ⫽ 16.25, P ⬍ 0.0001), and remained elevated at
168 h for both measures. Overall, the ECC and
AW⫺LO⫹ECC groups exhibited more swelling than the
Medicine & Science in Sports & Exercise姞
1895
FIGURE 1—Creatine kinase activity: Mean CK activity for all treatment groups at baseline (0 h) and 24, 48, 72, and 168 h after treatment.
*P < 0.05 compared with control group at same time point.
control group for both proximal (F4,38 ⫽ 7.09, P ⬍ 0.0002)
and distal (F4,38 ⫽ 4.12, P ⬍ 0.007) measures. Analysis of
a significant time-by-treatment interaction for both proximal
(F16,152 ⫽ 3.30, P ⬍ 0.0001) and distal (F16,152 ⫽ 1.96, P
⬍ 0.019) measures revealed that all experimental groups
exhibited a significant increase in circumference for at least
one time point (Fig. 4). Additionally, a difference between
experimental groups was observed at 24 h and 48 h, when
the AW⫺LO⫹ECC group exhibited a greater increase in
proximal swelling than the PW⫺LO⫹ECC group (P ⬍
0.05).
Muscle soreness. Peak soreness was noted at 48 h for
both extension and flexion, with no significant difference in
perception of soreness between the two measures (Table 1).
Soreness was elevated from baseline at 24, 48, and 72 h for
FIGURE 3—Changes in range of motion: Mean change in resting
extension (A) and active flexion (B) compared with baseline for all
treatment groups at 24, 48, 72, and 168 h after treatment. *P < 0.05
compared with control values.
both flexion (F4,152 ⫽ 51.71, P ⬍ 0.001) and extension
(F4,152 ⫽ 46.61, P ⬍ 0.0001), and all treatment groups
experienced significantly more soreness than the CONTROL group (F4,38 ⫽ 10.33 and 7.71, P ⬍ 0.0001 for
flexion and extension, respectively). Post hoc analyses of
significant time-by-treatment interactions for both flexion
and extension (F16,152 ⫽ 4.74 and 4.72, P ⬍ 0.0001) showed
that the high-heat passive warm-up before exercise group
exhibited significant change at the least number of time
points (Fig. 5).
DISCUSSION
FIGURE 2—Percent of maximal voluntary contraction: Mean percent
of MVC for all treatment groups at baseline (0 h) and 24, 48, 72, and
168 h after treatment. *P < 0.05 compared with control values.
1896
Official Journal of the American College of Sports Medicine
It was our goal to assess the effects of warm-up on
damage resulting from high-load eccentric muscle contraction. Although we observed differences with respect to how
our experimental groups differed from control values, the
evidence of this study does not support the hypothesis that
warm-up immediately preceding eccentric exercise results
http://www.acsm-msse.org
groups were below control values throughout the course of
the study, and a similar recovery pattern was evident between groups. Nosaka and Clarkson (22), using a similar
active warm-up protocol, observed no significant differences in strength loss at 24 h for those conducting eccentric
exercise with or without warm-up. Contrary to our results,
however, they did note that strength recovery was significantly faster in the warm-up condition. Rodenburg et al. (25)
reported no difference in strength loss or recovery when
warm-up preceded eccentric exercise, but they combined
warm-up with other treatments, making comparison with
our study difficult.
We also observed no significant differences in range of
motion, circumferential change, and subjective pain levels
between those groups that conducted warm-up and those
that did not. Again, Nosaka and Clarkson noted that pain
decreased and range of motion was improved when active
warm-up preceded eccentric exercise (22). They did, however, use a crossover design that allowed for comparison
between like extremities, which may have created better
control for variability in individual response, increasing
their ability to see statistical differences resulting from
treatment.
As our active warm-up was similar to that used by Nosaka
and Clarkson (22), our differing results may also reflect the
use of different damage protocols. In their study, subjects
performed 12 low-velocity eccentric contractions (40°·s⫺1)
FIGURE 4 —Arm circumference: Percent of proximal (A) and distal
(B) arm circumference compared with baseline for all treatment
groups at baseline (0 h) and 24, 48, 72, and 168 h after treatment. *P
< 0.05 compared with control values, **P < 0.05 compared with
PWⴚLOⴙECC.
in fewer clinical signs of muscle damage than eccentric
exercise alone.
It is well known that the response to eccentric exercise
varies widely among individuals (27). Although CK activity
is no exception and exhibits large intersubject variability, it
is a sensitive marker of muscle damage (8). Mean CK values
at 72 h for the AW⫺LO⫹ECC and ECC conditions in our
study (5199.4 ⫾ 8183.3 IU·L⫺1 and 2109.1 ⫾ 3331.3
IU·L⫺1, respectively) were greater than the values observed
by Nosaka and Clarkson at 72 h (22) (approximately 400 ⫾
200 IU·L⫺1 and 1800 ⫾ 1400 IU·L⫺1), who tested the
effects of active warm-up preceding eccentric exercise. Although this may represent increased damage, it may also
reflect normal individual variability.
Significant decreases in force production resulted from
our eccentric damage protocol, and the pattern of strength
loss and recovery was similar to that described in other
studies (22,25). Our warm-up conditions, however, had no
significant effect on force production. All experimental
WARM-UP BEFORE ECCENTRIC EXERCISE
FIGURE 5—Muscle soreness: Changes in subjective muscle soreness
during extension (A) and flexion (B) as noted on a 100-mm visual
analog scale at baseline (0 h) and 24, 48, 72, and 168 h after treatment.
* P < 0.05 compared with control values.
Medicine & Science in Sports & Exercise姞
1897
and exhibited an average force loss of 28% at 24 h. Our
subjects performed 50 maximal effort high-velocity eccentric contractions (120°·s⫺1) and exhibited an average force
loss of 39% at 24 h, similar to other studies using more
severe damage protocols (23,25). As injury has been observed to increase with the duration and velocity of eccentric
contraction (16), our protocol may have produced considerably more damage than the protocol used by Nosaka and
Clarkson, and may have worsened the clinical indicators of
muscle damage.
Further evidence that our protocol was too severe is the
number of “high responders” we observed. High responders
have been defined as individuals who present with pronounced swelling, extended decrements in muscle function,
and greatly elevated CK levels after eccentric exercise (27).
Sayers et al. (27) reported a 3% incidence of high responders (6/204) in subjects who performed 50 low-velocity eccentric contractions. Using the same loosely defined criteria,
we observed a 14% incidence of subjects who exhibited
pronounced swelling (2.3- to 3.9-cm increase in arm circumference), significant strength loss (34.4 –58.2% of baseline), or both. All exhibited greatly elevated CK levels
(4,832–10,483 IU·L⫺1) and were noted to have strength
decrements well beyond the 7-d study period, during which
time they were followed to assure swelling and strength loss
resolved. At least one subject exhibiting this high response
to eccentric exercise was observed in each treatment group.
Using an eccentric protocol that results in such severe
symptoms may not be appropriate in studying the efficacy of
clinical interventions for eccentric muscle damage. The
symptoms exhibited by high responders characterize rhabdomyolysis, a condition of muscle breakdown that can be
caused by excessive exercise (8). A less intense damageproducing protocol of 12 eccentric contractions (22) versus
the 50 –70 eccentric contractions used in this study and
others (29) may be more representative of what is seen
clinically when untrained individuals perform eccentric exercise. Further studies should be encouraged to define the
appropriate damage protocol to be used to produce a clinical
presentation of eccentric muscle damage. As well, it is
clinically important to define the criteria for a diagnosis of
rhabdomyolysis, a condition that could potentially lead to
adverse systemic response if large muscle groups are involved (8). Further research is warranted that can elucidate
the reasons certain individuals exhibit this exceptionally
high response.
Although there is ample evidence to suggest that muscle
warm-up of as little as 1°C might decrease strain injury
(13,21,26,28,30,31), our active and low-heat passive
warm-up treatments, which elevated tissue temperature by
approximately 1°C, may not have increased tissue extensibility by amounts great enough to observe clinically significant changes. Additionally, warm-up may increase extensibility when noncontracted muscle is stretched (21,30) but
might not affect extensibility when stretch is imposed on
contracted muscle. Indeed, warmed, noncontracted muscle
achieved increased elongation when warmed from 25° to
40°C, but an eccentrically lengthened muscle warmed by the
1898
Official Journal of the American College of Sports Medicine
same degree did not achieve greater elongation at the point
of failure (21). The increased elongation noted in warmed,
noncontracted muscle during stretch may have occurred
within the contractile element, rather than in the collagenous
connective tissue (15). During eccentric lengthening, contracting sarcomeres may prevent elongation from occurring
within the contractile apparatus.
Further, it has been reported that tissue temperature elevation of at least 3– 4°C is required to increase connective tissue
extensibility (13). Interestingly, elevated tissue temperature has
also been hypothesized to increase eccentric muscle damage by
enhancing tissue degradation (1). Indeed, increasing mouse
muscle temperature from 25 to 35°C resulted in greater damage
during eccentric contraction (33). Risk factors for exerciseinduced rhabdomyolysis include high ambient temperature,
high humidity, and lack of proper hydration, all of which may
result in elevated body temperature (8). For this reason, we
limited our high-heat treatment group (PW⫺HI⫹ECC) to a
small number of volunteers (N ⫽ 4).
We observed no evidence that passively elevating muscle
temperature by 3– 4°C before eccentric contractions caused
more damage than passive warm-up of ⬇1°C or eccentric
exercise alone. Although this may appear to contradict the
results of Zerba and Faulkner (33), there were major differences in our methods; they induced a 10°C rise in temperature
of mouse muscle (from 25 to 35°C), whereas we heated human
muscle by 3– 4°C (from approximately 37 to 41°C). To provide
more information, three of the four subjects were treated with
the eccentric exercise without warm-up condition 3 months
later using the contralateral arm. We observed no differences in
indicators of muscle damage. Although we cannot conclude
that passively heating muscle tissue by 3– 4°C prevents clinical
presentation of muscle damage, it does appear that local heating of muscle tissue may not exacerbate muscle damage in
human subjects. In fact, our PW⫺HI⫹ECC group exhibited
significant change from control values at the least number of
time points for measures of range of motion and pain. Due to
the small sample size of our PW⫺HI⫹ECC group, however,
we can only draw preliminary conclusions as to the effect of
localized heat on eccentric muscle damage. Further study with
more subjects is necessary before definitive conclusions can be
drawn.
The ECC and AW⫺LO⫹ECC groups exhibited a greater
percent change in circumference than controls. Additionally,
we observed significant differences in response between active
and passive warm-up with the proximal arm circumference
measure, which was significantly greater at 24 h and 48 h in the
ECC⫹AW⫺LO group than the ECC⫹PW⫺LO group. Although studies have observed increased tissue extensibility
after both active and passive warm-up (21,26), it has been
suggested that contractile activity during active warm-up may
result in higher myotatic feedback loop activation and increased stiffness, and may impose limitations on fiber elongation (10), increasing the chance of strain damage. Given an
identical lengthening stimulus, elevating tissue temperature
passively before stretch is thought to decrease strain damage by
decreasing the sensitivity of the muscle spindles to stretch (11).
This hypothesized increase in tissue stiffness resulting from
http://www.acsm-msse.org
active warm-up may have resulted in the increased swelling
we observed in our active warm-up group. This is not
conclusive, however, as other indicators of damage were not
statistically different between those who performed active
warm-up and those who performed passive warm-up before
eccentric exercise.
CONCLUSION
than active warm-up or no warm-up in attenuating swelling
but does not support the use of warm-up to prevent, attenuate, or resolve more quickly the strength loss, loss of
motion, or soreness resulting from eccentric muscle damage
as produced in this study. Additional research on the use of
warm-up techniques is encouraged to provide scientific rationale for those who regularly use warm-up techniques to
achieve specific performance or injury-prevention goals.
Our observations suggest that passive warm-up performed before eccentric exercise may be more beneficial
We gratefully acknowledge the students at Brigham Young University who volunteered for this study.
REFERENCES
1. ARMSTRONG, R. B., G. L. WARREN, and J. A. WARREN. Mechanisms
of exercise-induced muscle fibre injury. Sports Med. 12:184 –207,
1991.
2. BROWN, S., S. DAY, and A. DONNELLY. Indirect evidence of human
skeletal muscle damage and collagen breakdown after eccentric
muscle actions. J. Sports Sci. 17:397– 402, 1999.
3. BYRD, S. K. Alterations in the sarcoplasmic reticulum: a possible
link to exercise-induced muscle damage. Med. Sci. Sports Exerc.
24:531–536, 1992.
4. DAVIES, C. T. M., and K. YOUNG. Effect of heating on the contractile properties of triceps surae and maximal power output
during jumping in elderly men. Gerontologist 31:1–5, 1985.
5. DRAPER, D. O., K. KNIGHT, T. FUJIWARA, and J. C. CASTEL. Temperature change in human muscle during and after pulsed shortwave diathermy. J. Orthop. Sports Phys. Ther. 29:13–22, 1999.
6. DRAPER, D. O., S. SCHULTHIES, P. SORVISTO, and A. M. HAUTALA.
Temperature changes in deep muscles of humans during ice and
ultrasound therapies: an in vivo study. J. Orthop. Sports Phys.
Ther. 21:153–157, 1995.
7. GARRETT, W. E., M. R. SAFRAN, A. V. SEABER, R. R. GLISSON, and
B. M. RIBBECK. Biomechanical comparison of stimulated and
nonstimulated skeletal muscle pulled to failure. Am. J. Sports Med.
15:448 – 454, 1987.
8. HAMER, R. When exercise goes awry: exertional rhabdomyolysis.
South. Med. J. 20:548 –551, 1997.
9. HAN, X. Y., W. WANG, J. KOMULAINEN, et al. Increased mRNAs for
procollagens and key regulating enzymes in rat skeletal muscle
following downhill running. Pflugers Arch. 437:857– 864, 1999.
10. HARDY, L., R. LYE, and A. HEATHCOTE. Active versus passive
warm-up regimes and flexibility. Carnegie Res. Papers 1:23–30,
1990.
11. KULUND, D. N., and M. TOTTOSSY. Warm-up, strength and power.
Orthop. Clin. North Am. 4:427– 448, 1983.
12. LEHMANN, J. Therapeutic Heat and Cold, 4th Ed. Baltimore: Williams & Wilkins, 1990.
13. LEHMANN, J. F., A. J. MASOCK, C. G. WARREN, and J. N. KOBLANSKI.
Effect of therapeutic temperatures on tendon extensibility. Arch.
Phys. Med. Rehabil. 51:481– 487, 1970.
14. MACPHERSON, P. C., M. A. SCHORK, and J. A. FAULKNER. Contraction-induced injury to single fiber segments from fast and slow
muscles of rats by single stretches. Am. J. Physiol. 271:C1438 –
C1446, 1996.
15. MAGID, A., and D. J. LAW. Myofibrils bear most of the resting
tension in frog skeletal muscle. Science 230:1280 –1282, 1985.
16. MCCULLY, K. K., and J. A. FAULKNER. Characteristics of lengthening contractions associated with injury to skeletal muscle fibers.
J. Appl. Physiol. 61:293–299, 1986.
17. MCNEIL, P. L., and R. KHAKEE. Disruptions of muscle fiber plasma
membranes. Role in exercise-induced damage. Am. J. Pathol.
140:1097–1109, 1992.
WARM-UP BEFORE ECCENTRIC EXERCISE
18. MORGAN, D. L. New insights into the behavior of muscle during
active lengthening. Biophys. J. 57:209 –221, 1990.
19. MUMMERY, W. K., J. C. SPENCE, J. A. VINCENTEN, and D. C.
VOAKLANDER. A descriptive epidemiology of sport and recreation
injuries in a population-based sample: results from the Alberta
Sport and Recreation Injury Survey (ASRIS). Can. J. Public
Health 89:53–56, 1998.
20. NEWHAM, D. J., G. MCPHAIL, K. R. MILLS, and R. H. T. EDWARDS.
Ultrastructural changes after concentric and eccentric contractions
of human muscle. J. Neurol. Sci. 61:109 –122, 1983.
21. NOONAN, T. J., T. M. BEST, A. V. SEABER, and W. E. GARRETT.
Thermal effects on skeletal muscle tensile behavior. Am. J. Sports
Med. 21:517–522, 1993.
22. NOSAKA, K., and P. CLARKSON. Influence of previous concentric
exercise on eccentric exercise-induced muscle damage. J. Sports
Sci. 15:477– 483, 1997.
23. NOSAKA, K., and P. M. CLARKSON. Changes in indicators of inflammation after eccentric exercise of the elbow flexors. Med. Sci.
Sports Exerc. 28:953–961, 1996.
24. ROBERGS, R. A., D. D. PASCOE, D. L. COSTILL, et al. Effects of
warm-up on muscle glycogenolysis during intense exercise. Med.
Sci. Sports Exerc. 23:37– 43, 1991.
25. RODENBURG, J. B., D. STEENBEEK, P. SCHIERECK, and P. R. BAR.
Warm-up, stretching and massage diminish harmful effects of
eccentric exercise. Int. J. Sports Med. 15:414 – 419, 1994.
26. SAFRAN, M. R., W. E. GARRETT, A. V. SEABER, R. R. GLISSON, and
B. M. RIBBECK. The role of warm-up in muscular injury prevention. Am. J. Sports Med. 16:123–129, 1988.
27. SAYERS, S. P., P. M. CLARKSON, P. A. ROUZIER, and G. KAMEN.
Adverse events associated with eccentric exercise protocols: six
case studies. Med. Sci. Sports Exerc. 31:1697–1702, 1999.
28. SHELLOCK, F. G., and W. E. PRENTICE. Warming-up and stretching
for improved physical performance and prevention of sportsrelated injuries. Sports Med. 2:267–278, 1985.
29. STAUBER, W. T., P. M. CLARKSON, V. K. FRITZ, and W. J. EVANS.
Extracellular matrix disruption and pain after eccentric muscle
action. J. Appl. Physiol. 69:868 – 874, 1990.
30. STRICKLER, T., T. MALONE, and W. E. GARRETT. The effects of
passive warming on muscle injury. Am. J. Sports Med. 18:141–
145, 1990.
31. WARREN, C. G., J. F. LEHMANN, and J. N. KOBLANSKI. Elongation of
rat tail tendon: effect of load and temperature. Arch. Phys. Med.
Rehabil. 52:465– 474, 1971.
32. WIRTH, V. J., B. L. VANLUNEN, D. MISTRY, E. SALIBA, and F. C.
MCCUE. Temperature changes in deep muscles of humans during
upper and lower extremity exercise. J. Athletic Training 33:211–
215, 1998.
33. ZERBA, E., and J. A. FAULKNER. At different muscle temperatures,
contraction-induced injury correlates with power absorption.
FASEB J. 4:A815, 1990.
Medicine & Science in Sports & Exercise姞
1899